Screening to Prevent Cervical Cancer: Guidelines for the Management of Asymptomatic Women with Screen Detected Abnormalities

Transcription

1 Screening to Prevent Cervical Cancer: Guidelines for the Management of Asymptomatic Women with Screen Detected Abnormalities APPROVED BY

2 Screening to Prevent Cervical Cancer: Guidelines for the Management of Asymptomatic Women with Screen Detected Abnormalities APPROVED BY THE NHMRC ON 9 JUNE 2005

3 ISBN: Paper-based publications Commonwealth of Australia 2005 This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the Commonwealth. Requests and inquiries concerning reproduction and rights should be addressed to the Commonwealth Copyright Administration, Attorney General s Department, Robert Garran Offices, National Circuit, Barton ACT 2600 or posted at Internet sites Commonwealth of Australia 2005 This work is copyright. You may download, display, print and reproduce this material in unaltered form only (retaining this notice) for your personal, non-commercial use or use within your organisation. Apart from any use as permitted under the Copyright Act 1968, all other rights are reserved. Requests and inquiries concerning reproduction and rights should be addressed to Commonwealth Copyright Administration, Attorney General s Department, Robert Garran Offices, National Circuit, Barton ACT 2600 or posted at These guidelines were approved by the National Health and Medical Research Council (NHMRC) at its 157th session on 9 June 2005, under section 14A of the National Health and Medical Research Council Act Approval by the NHMRC is granted for a period not exceeding five years, at which date the approval expires. The NHMRC expects that all guidelines will be reviewed no less than once every five years. Disclaimer This document is a general guide to appropriate practice, to be followed subject to the clinician s judgment and the patient s preference in each individual case. These guidelines are designed to provide information to assist decision making and are based on the best evidence available at the time of publication. The development of these guidelines was facilitated by the NSW Cervical Screening Program on behalf of the National Cervical Screening Program. These guidelines can be downloaded from the NHMRC website: Editing and formatting by Biotext Pty Ltd, Canberra

4 Preamble In approving this document, Screening to prevent cervical cancer: guidelines for the management of asymptomatic women with screen detected abnormalities, at its 157th session on 9 June 2005, the National Health and Medical Research Council expressed its support for the National Cervical Screening Program as an important way to maintain the health of all Australian women. The new guidelines will not only give medical practitioners evidence-based recommendations to better manage patients who have an abnormal Pap smear, but will, through the planned implementation and education programs, raise women s awareness of the usefulness of Pap smears as a tool in preventing deaths from cervical cancer. The National Health and Medical Research Council, however, recommended that the screening interval in Australia be reviewed as soon as possible to ensure that the National Cervical Screening Program is consistent with international best practice. National Health and Medical Research Council 9 June 2005 Preamble i

5

6 Foreword About two million Australian women have a Pap test for cervical cancer screening each year. The clinical management of women with abnormal Pap test results involves health professionals across a broad spectrum of disciplines and sectors. These guidelines have been developed to assist women and health professionals to achieve the best outcomes. The target audience includes primary health care providers, general practitioners, cytopathologists, gynaecologists and gynaecological oncologists, as well as public health service administrators. The previous National Health and Medical Research Council (NHMRC) guidelines Screening to Prevent Cervical Cancer: Guidelines for the Management of Women with Screen Detected Abnormalities were developed by an expert group convened by the NHMRC. They were endorsed by the NHMRC in December 1993 and published in The guidelines were based on the latest available literature at that time and provided effective guidance to practitioners on that basis. However, they were not formulated in line with recent NHMRC standards for clinical practice guidelines. In 2001, the National Advisory Committee to the National Cervical Screening Program (NCSP) of the Australian Department of Health and Ageing requested that the guidelines be reviewed and updated. The New South Wales Cervical Screening Program funded and managed the review under the direction of the NCSP. The review has been undertaken according to the NHMRC requirements described in A Guide to the Development, Implementation and Evaluation of Clinical Practice Guidelines (NHMRC 1999), and the subsequent handbook series on preparation of clinical practice guidelines (NHMRC 2000abc, 2001, 2002). These updated guidelines address the current state of cervical cancer in Australia, the natural history of the disease and terminology for cervical cytology; management of squamous abnormalities, glandular abnormalities and special clinical circumstances; and psychosocial, economic and implementation issues. The development of these guidelines has involved widespread consultation with all relevant professional bodies and a wide range of clinicians and consumers. However, while the guidelines are based on good population data, it is important to note that they are only a guide to clinical practice. Clinicians must make individual decisions in consultation with their patients based on individual clinical circumstances. The guidelines do not address issues related to routine screening interval or frequency, or give detailed information about the treatment of invasive cervical cancer. Furthermore, the guidelines explicitly exclude symptomatic women. The Royal Australian and New Zealand College of Obstetricians and Gynaecologists has issued management guidelines for women with intermenstrual and postcoital bleeding (RANZCOG 2002), which take precedence over these guidelines for such cases. The updated guidelines are based on revised terminology for cervical cytology reporting in Australia, which will be known as the Australian Modified Bethesda System 2004 (AMBS 2004). The previous NHMRC-endorsed terminology was a uniquely modified version of the The Bethesda System 1991 (TBS 1991), which was developed in the United States. However, its numerical categorisations of abnormalities implied cervical intraepithelial neoplasia (CIN) progression from CIN 1 to CIN 2 to CIN 3 and then cancer. We now know that this is not an accurate reflection of the biology of human Foreword iii

7 papillomavirus (HPV), cervical cancer and its precursors. During the past decade there has been a greater understanding of the natural history of cervical cancer. The nature of CIN 1 as an infective process rather than a neoplastic one has been demonstrated by recent work in molecular biology and epidemiology. These advances suggest that, rather than an inevitable linear progression towards cancer, most HPV infections acquired by women resolve without medical intervention. Women with low-grade epithelial abnormalities on cervical cytology are currently the subject of a number of clinical trials investigating their optimal management. However, applying international research to the Australian context has, until now, been extremely difficult because of the structural differences in the terminology system used. The revised terminology system, AMBS 2004, which is presented in this document, reflects a modern understanding of HPV infection, cervical cancer and its precursors and is compatible with terminology systems now used internationally. The Australian Government Department of Health and Ageing will fund and take responsibility for the dissemination, implementation and monitoring of these guidelines, once they are approved by the NHMRC. A plan for implementing the guidelines is included in Appendix 4 with recommendations for further review. This includes provision for a consumer guide and guides for general practitioners. These new guidelines are a distillation of the very latest research and data, brought together by some of the leading experts in this field, both internationally and in Australia. I believe they will result in a significant improvement in the care and treatment of asymptomatic women with screen-detected cervical abnormalities in Australia, and I strongly commend them to you. Professor Ian Hammond Chair, Guidelines Review Group Clinical Professor School of Women s and Infants Health University of Western Australia iv Screening to prevent cervical cancer

8 Contents Preamble Foreword...i... iii Summary of guidelines...xi 1 Guidelines review process Introduction Literature review Assessment and classification of evidence Cervical cancer in Australia Current concepts of HPV infection and the natural history of cervical neoplasia HPV infection causes cancer of the cervix Cervical cancer is a rare outcome of HPV infection Virology of HPV infection New evidence about the natural history of LSIL Natural history of HSIL Terminology Background Australian Modified Bethesda System 2004 (AMBS 2004) Explanation and definition of AMBS 2004 terminology Preparation of cervical cytology reports using AMBS Review of cervical smear data using previous and new Australian terminology Management of women with unsatisfactory Pap smears Background Australian definition of unsatisfactory Pap smear Application Management of unsatisfactory Pap test results Management of low-grade squamous abnormalities Introduction Australian evidence Why are cancers diagnosed after low-grade cytology reports?...33 Contents v

9 6.4 Management options Management of a cytological prediction of possible or definite LSIL Subsequent management Fluctuating abnormalities Early referral for colposcopy in exceptional circumstances What evidence is available about the safety of cytological surveillance for women with low-grade cytology? Estimate of net effect of these guidelines on incidence of cancer Balancing the risks of investigation and treatment with the risk of cervical cancer Safety monitoring Why is repeat cytology at 12 months safe for Australian women with LSIL cytology? Colposcopic assessment and management of a woman with a cytological prediction of LSIL Management of high-grade squamous abnormalities Introduction Evaluation of women with high-grade squamous abnormalities on cytology report Histological confirmation Management and treatment of women with histological diagnoses of high-grade squamous intraepithelial lesions Further issues in the colposcopic assessment of women with possible high-grade squamous lesions Follow-up of women treated for high-grade squamous intraepithelial lesions Management of cervical glandular abnormalities Introduction Frequency of cervical cytology reports of glandular abnormalities Natural history of cervical adenocarcinoma Initial evaluation and referral of women with glandular abnormalities on cervical cytology Role of colposcopy, target biopsy, cone biopsy and endocervical curettage in the assessment of glandular lesions Further clinical management of women with glandular abnormalities Controversies in the management of women with histological evidence of AIS Follow-up of women treated for AIS...71 vi Screening to prevent cervical cancer

10 8.9 Endometrial and other malignancies Special clinical circumstances Pregnancy Immunosuppressed women Normal endometrial cells in postmenopausal women Women exposed in utero to diethylstilboestrol Other issues Psychosocial issues Economic considerations Introduction Changes to terminology Low-grade squamous abnormalities Glandular abnormalities HPV testing...85 Appendix 1 Guidelines Review Group membership...87 Appendix 2 Process for development of guidelines...91 Appendix 3 Appendix 4 Appendix 5 Appendix 6 Appendix 7 Appendix 8 Appendix 9 Methods...97 Dissemination, implementation, monitoring and review of guidelines Examples of reports conforming to the terminology requirements Principles of screening Outcome after a cytological prediction of low-grade abnormality in Outcome after a cytological prediction of glandular abnormality in RANZCOG guidelines for referral for investigation of intermenstrual and postcoital bleeding Appendix 10 Analysis of cancer outcomes from recommended management of women with LSIL/possible LSIL cytology Appendix 11 Evaluation of low-grade guidelines against historical data from Victoria Contents vii

11 Appendix 12 Safety monitoring of the recommended management for women with low-grade cervical cytology Appendix 13 Safety monitoring of the recommended management of women with treated high-grade intraepithelial abnormalities Abbreviations Glossary References Tables Table 1.1 NHMRC levels of evidence...2 Table 1.2 Dimensions of evidence...2 Table 1.3 Designation of evidence used in these guidelines...3 Table 2.1 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 6.1 Incidence of cervical cancer and mortality rate, selected countries, Regression and progression rates and times for LSIL, Brazilian longitudinal study...13 Cytological regression by time and degree of HSIL, Netherlands natural history study...16 Size of CIN 3 in the presence of microcarcinoma and for varying grades of CIN on cytology, cone biopsy study...17 Total dimension score (TDS) for CIN 3 at exit visit, ALTS trial...17 Comparison of the Australian Modified Bethesda System (AMBS 2004) with previous Australian terminology and The Bethesda System (TBS 2001)...23 Royal College of Pathologists of Australasia Quality Assurance Program correlations...26 Kappa scores for reassessment of slides...27 Positive predictive value and outcomes of CIN 2 or 3 prediction...27 Prevalence of squamous intraepithelial lesions in NSW women, by age...32 Table 6.2 Detection of CIN 3 after ASCUS in ALTS...41 Table 6.3 Detection of CIN 3 after LSIL in ALTS...42 Table 6.4 Sample size estimations (P < 0.05, 2-sided, power 90%)...42 Table 6.5 Recommended management for women with abnormal cytology, by country...44 Table 6.6 Vital statistics for cervical neoplasia for Finland and Victoria...45 viii Screening to prevent cervical cancer

12 Table A2.1 Table A7.1 Table A7.2 Table A7.3 Table A7.4 Table A7.5 Table A7.6 Table A7.7 Table A7.8 Search protocols for identifying systematic reviews and randomised controlled trials Proportion of women with LGEA cytology reports, by age and type of LGEA report, Proportion of women with LGEA cytology reports, by region and type of LGEA report, Proportion of women with no further information, by age and type of LGEA report Proportion of women undergoing cervical biopsy within six months, by age and type of LGEA report Profile of histology results within six months, by age and type of LGEA report Profile of histology results within 6 24 months, by age and type of LGEA report Profile of worst cytology results within 0 24 months, by age and type of LGEA report Composite results over the 24-month period (number and proportion) Table A7.9 Composite results over the 24-month period (rates per 1000 women) Table A7.10 Screening history of women diagnosed with microinvasive cancer after an LGEA report Table A7.11 Criteria for the selection of women for each registry Table A8.1 Table A8.2 Table A8.3 Table A8.4 Table A8.5 Table A8.6 Table A8.7 Table A8.8 Proportion of women with glandular abnormalities on cytology, by region of Australia, Profile of histology results performed within six months, by degree of cytological abnormality Profile of histology results performed within 6 24 months of the index smear, by degree of cytological abnormality Profile of worst cytology results performed within 0 24 months of the index smear, by degree of cytological abnormality Composite results over the 24-month period Profile of cervical cancers diagnosed among women in the study Screening history of women diagnosed with cervical cancer after a minor (glandular) atypia report Criteria for the selection of women for each registry Table A10.1 Sensitivity analysis Table A11.1 Evaluation of 2004 guidelines against women diagnosed with cervical cancer in Victoria during 1999, 2000 and Contents ix

13 Figures Figure 2.1 Age-specific incidence rates of cervical cancer, Australia, Figure 2.2 Age-standardised incidence rates for cervical cancer, by histological type, for women aged years, Australia, Figure 3.1 Natural history of HPV infection and cervical cancer precursors...11 Figure 3.2 Figure 6.1 Figure 6.2 Figure 7.1 Figure 7.2 Figure 8.1 Frequency of histologic CIN 2 and CIN 3 per 100,000 screened women versus incidence of cervical cancer per 100,000 unscreened women...15 Management of a cytological prediction of possible or definite LSIL...39 Age-specific incidence rate for cervical cancer for Australia, the Netherlands and Finland...45 Management of women with cervical cytology predicting possible high-grade squamous lesions...58 Frequency of positive HPV testing, by post-treatment follow-up period...60 Management of women with invasive adenocarcinoma suggested on cervical cytology or biopsy...68 Figure 8.2 Management of women with AIS on cervical cytology...69 Figure 8.3 Figure 9.1 Management of women with cervical cytology suggesting possible high-grade glandular lesions...70 Management of a high-grade lesion in pregnancy...75 x Screening to prevent cervical cancer

14 Summary of guidelines Management of women with unsatisfactory Pap smears GUIDELINE Unsatisfactory Pap test reports A woman with an unsatisfactory Pap test report should have a repeat smear in 6 12 weeks, with correction, when possible, of the problem that caused the unsatisfactory smear. EVIDENCE Consensus Management of low-grade squamous abnormalities Human papilloma virus (HPV) testing There is insufficient evidence to support the use of HPV testing in the triage of low-grade squamous intraepithelial lesions. Index Pap test report of low-grade squamous intraepithelial lesions (LSIL) A woman with a Pap test report of LSIL should be managed in the same way irrespective of whether the abnormality is regarded as possible or definite and should be recommended for a repeat Pap test in 12 months. Index Pap test reports of LSIL in women aged 30+ years A woman aged 30 years or more with a Pap test report of LSIL, without a history of negative smears in the preceding two to three years, should be offered either immediate colposcopy or a repeat Pap smear within six months. Twelve-month repeat Pap test after index test results of LSIL If the 12-month repeat Pap test is reported as showing high-grade changes (definite or possible), the woman should be referred for colposcopic assessment. Any woman whose repeat Pap test at 12 months is again reported as showing changes suggestive of LSIL (whether possible or definite), should be referred for colposcopic assessment. MSAC 2002 Australian registry data; Level III-2: three cohort studies of clearance interval (Ho et al 1998, Moscicki et al 1998, Woodman et al 2001) Australian registry data Level IV (Schoolland et al 1998, Sparkes et al 2000, Performance Standards 2003) Level III-2: three cohort studies of clearance interval (Ho et al 1998, Moscicki et al 1998, Woodman et al 2001) Summary of guidelines xi

15 If the 12-month repeat Pap test is reported as normal, the woman should have a further repeat Pap test in 12 months (ie 24 months after the index smear). Fluctuating repeat Pap test results Referral for colposcopy should be considered for a woman if she has two LSIL/possible LSIL reports (at least 12 months apart) within a 3-year timeframe, regardless of intervening normal cytology reports. Colposcopic assessment of women with Pap test reports of LSIL If, at colposcopy, a high-grade lesion is seen or suspected, targeted biopsy should be performed for histological confirmation before definitive therapy. If the colposcopic assessment is normal, the woman should be referred back for annual cytological surveillance until two normal smears are obtained, and then resume routine screening according to the recommendation for the average population. If the colposcopic assessment is satisfactory and a low-grade lesion is suspected, target biopsy can be performed to confirm this diagnosis. Treatment of histologically confirmed low-grade squamous lesions is not recommended, as such lesions are considered to be an expression of a productive HPV infection. Histologically confirmed low-grade squamous abnormalities can be safely managed by repeat cytology at 12 and 24 months. If both smears are negative, it is recommended that the woman return to screenings at the intervals recommended for the average woman. If either repeat smear shows possible or definite LSIL, the woman should be advised to continue having annual smears until at least two are negative, at which time she can return to routine screening. If the colposcopic assessment is unsatisfactory, consideration should be given to repeating the Pap test in 6 12 months. In asymptomatic women and in the absence of any cytologic, colposcopic or histologic suggestion of high-grade disease, further diagnostic procedures, such as cone biopsy or loop excision, are not indicated. Level III-2: three cohort studies of clearance interval (Ho et al 1998, Moscicki et al 1998, Woodman et al 2001) Consensus Consensus (RANZCOG 2001) Consensus (RANZCOG 2001) Consensus (RANZCOG 2001) Consensus (RANZCOG 2001) Consensus (RANZCOG 2001) xii Screening to prevent cervical cancer

16 Management of high-grade squamous abnormalities Referral of women with Pap test reports of possible high-grade squamous lesions A woman with a Pap test report of possible high-grade squamous lesion should be referred to a gynaecologist for colposcopic assessment and targeted biopsy where indicated. Referral of women with Pap test reports of high-grade squamous intraepithelial lesions (HSIL) A woman with a Pap test report of HSIL should be referred to a gynaecologist for colposcopic assessment and targeted biopsy where indicated. Referral of women with Pap test reports of HSIL with additional features suggestive of an invasive component A woman with a Pap test report of HSIL, with additional features suggestive of an invasive component, should be referred to a gynaecologist with expertise in colposcopic evaluation of suspected gynaecological malignancies or to a gynaecological oncologist, ideally within two weeks. Referral of women with Pap test reports of SCC A woman with a Pap test report of SCC should be referred to a gynaecological oncologist or to a gynaecological oncology unit for urgent evaluation, ideally within two weeks. Histological confirmation Histological confirmation of a high-grade lesion is required before definitive treatment is undertaken. See and treat is not recommended. Treatment of a high-grade squamous intraepithelial abnormality Women with a histological diagnosis of CIN 2 or CIN 3 should be treated in order to reduce the risk of developing invasive cervical carcinoma. Fertility-sparing treatments Local ablative or excisional treatments should destroy or remove tissue to a depth of at least 7 mm. There is no clearly superior method of fertility-sparing treatment for CIN 2 and 3. Level IV (Schoolland et al 1998, Sparkes et al 2000, VCCR 2002) Level IV (VCCR 2002, Sparkes et al 2000) Consensus Consensus Consensus Consensus Level III-2 (Östör 1993b) Level IV (Burke 1982, Jordan et al 1985) Level I (Martin-Hirsch et al 2000) Summary of guidelines xiii

17 Ablative therapy Ablative therapy may be considered, provided: 1. The cervix has been assessed by an experienced colposcopist. 2. A targeted biopsy has confirmed the diagnosis. 3. There is no evidence of an invasive cancer on cytology, colposcopic assessment or biopsy. 4. The entire cervical transformation zone has been visualised. 5. There is no evidence of a glandular lesion on cytology or biopsy. Cryotherapy for treatment of CIN 3 It is advisable that women with CIN 3 are not treated with cryotherapy. Loop electro-excisional procedure (LEEP) Excess diathermy artefact should be avoided when using LEEPs in order to allow comprehensive pathological examination, including margin status. Cone biopsy Cone biopsy may be necessary to treat women with high-grade squamous lesions and absolute indications that include: 1. failure to visualise the upper limit of the cervical transformation zone in a woman with a high-grade squamous abnormality on her referral cervical smear (ie unsatisfactory colposcopy) 2. suspicion of an early invasive cancer on cytology, biopsy or colposcopic assessment 3. the suspected presence of an additional significant glandular abnormality (ie adenocarcinoma in situ) on cytology or biopsy (ie a mixed lesion). Careful attention should be paid to tailoring treatment to the individual woman, taking into account the size, extent, situation and severity of the lesion. Consensus Level IV (Anderson and Husth 1992) Consensus Consensus Consensus xiv Screening to prevent cervical cancer

18 Management of women previously treated for HSIL A woman previously treated for HSIL requires a colposcopy and cervical cytology at 4 6 months after treatment. Cervical cytology and HPV typing should then be carried out at 12 months after treatment and annually thereafter until the woman has tested negative by both tests on two consecutive occasions. The woman should then be screened according to the recommendation for the average population. A woman already undergoing annual cytological review for follow-up of a previously treated HSIL, as advised by the previous NHMRC guidelines (1994), may be offered HPV testing as described above. Once she has tested negative by both cytology and HPV typing on two consecutive occasions, she should be screened according to the recommendation for the average population. Level IV (Chua and Hjerpe 1997, Bollen et al 1999, Jain et al 2001, Lin et al 2001, Nobbenhuis et al 2001b, Paraskevaidis et al 2001, Bar-Am et al 2003, Zielinski et al 2003, Chao et al 2004) Level IV (Chua and Hjerpe 1997, Bollen et al 1999, Jain et al 2001, Lin et al 2001, Nobbenhuis et al 2001b, Paraskevaidis et al 2001, Bar-Am et al 2003, Zielinski et al 2003, Chao et al 2004) Management of cervical glandular abnormalities Referral of women with Pap test reports of adenocarcinoma A woman with a Pap test report of adenocarcinoma of endometrial origin should be referred to a gynaecologist with expertise in the colposcopic evaluation of suspected malignancies or to a gynaecological oncologist. A woman with a cytological prediction of adenocarcinoma of either endocervical, extrauterine or unspecified origin should be referred to a gynaecological oncologist or a gynaecological oncology unit. Referral of women with Pap test reports of endocervical adenocarcinoma in situ (AIS) A woman with a Pap test report of endocervical AIS should be referred to a gynaecologist with expertise in the colposcopic evaluation of suspected malignancies or to a gynaecological oncologist Referral of women with Pap test reports of possible high-grade glandular lesions A woman with a Pap test report of possible high-grade glandular lesions should be referred to a gynaecologist with expertise in the colposcopic evaluation of suspected malignancies or to a gynaecological oncologist. Level III-3 (Mitchell et al 1993) Consensus Australian registry data Australian registry data Summary of guidelines xv

19 Referral of women with Pap test reports of atypical glandular or endocervical cells of undetermined significance A woman with a Pap test report of atypical glandular or endocervical cells of undetermined significance should be referred to a gynaecologist with expertise in the colposcopic evaluation of suspected malignancies. Colposcopic assessment of glandular lesions Colposcopic assessment is mandatory in the presence of a cervical cytology suggesting a glandular lesion. Cone biopsy for the assessment of glandular lesions Cold-knife cone biopsy should be considered the gold standard for the assessment of glandular lesions Referral of women with adenocarcinoma on cone or punch biopsy Women found to have invasive adenocarcinoma on cone or punch biopsy should be referred to a gynaecological oncologist or a gynaecological oncology unit for subsequent management. Management of women with a Pap test report of AIS If invasive carcinoma is not identified at colposcopic assessment, a cone biopsy should be undertaken. Hysterectomy should not be undertaken without prior cone biopsy to exclude invasive carcinoma. Management of women with AIS The management of women diagnosed with AIS on cone biopsy will be dependent upon the age and fertility requirements of the women and the status of excision margins. Hysterectomy is recommended for women who have completed childbearing because of the difficulties of reliable cytological follow-up, a high recurrence rate and the reported multifocality of the disease. Australian registry data Consensus Consensus Consensus Consensus Level IV (Cullimore et al 1992, Hopkins et al 1988, Muntz et al 1992, Im et al 1995, Poynor et al 1995, Denehy et al 1997 Widrich et al 1996, Wolf et al 1996, Azodi et al 1999, Hopkins 2000, Souter et al 2001, Anderson and Nielson 2002, Kennedy and Biscotti 2002, Shin et al 2002) xvi Screening to prevent cervical cancer

20 Special clinical circumstances Evaluation of an abnormal Pap test during pregnancy Women with low-grade cytologic lesions should be managed in the same way as for women with low-grade squamous abnormalities, with a repeat smear after 12 months. Women with high-grade lesions should be referred for colposcopic evaluation. Colposcopy during pregnancy The main aim of colposcopy in the pregnant woman is to exclude the presence of invasive cancer and to reassure the woman that her pregnancy will not be affected by the presence of an abnormal Pap test. Biopsy of the cervix is usually unnecessary in pregnancy, unless invasion is suspected colposcopically. Treatment of a high-grade lesion during pregnancy Definitive treatment of a high-grade lesion, with the exception of invasive cancer, may be deferred safely until after the pregnancy. Immunosuppressed women If an immunosuppressed woman has a screen-detected abnormality she should be referred for colposcopy, even if the lesion is low-grade, as cytological surveillance alone may be inadequate. Assessment and treatment should be by an experienced colposcopist. The whole of the lower genital tract will need evaluation as the same risk factors apply for cervical, vaginal, and vulval and perianal lesions. Treatment of the cervix should be by excisional methods. Follow-up after treatment should include colposcopy as well as cytology. Follow-up should be annual and indefinite. Level IV (Coppola et al 1997, Jain et al 1997, Woodrow et al 1998, Nguyen et al 2000) Level IV (Coppola et al 1997, Woodrow et al 1998, Nguyen et al 2000, Palle et al 2000) Level IV (Woodrow et al 1998) Level IV (Woodrow et al 1998, Palle et al 2000) Level IV Guerra et al (1998), Economos et al (1993) Level I/II (Sillman et al 1997, Spitzer 1999) Level III-1 (Petry et al 1994) Level III-1 (Petry et al 1994) Level I/II (Spitzer 1999) Level III-2 (Cordiner et al 1980) Level III-2 (Cordiner et al 1980) Summary of guidelines xvii

21 Postmenopausal women with normal endometrial cells Normal endometrial cells occurring in the Pap smear of an asymptomatic postmenopausal woman should not be reported. A symptomatic postmenopausal woman requires investigation irrespective of her Pap test status. Level III-2 (Gondos and King 1977, Gomez-Fernandez et al 2000, Ashfaq et al 2001, Montz 2001, Chang et al 2001, Brogi et al 2002) Level III-2 (RANZCOG 2002) Women exposed to diethylstilboestrol (DES) in utero DES-exposed women should be offered annual cytological screening and colposcopic examination of both the cervix and vagina. Screening should begin any time at the woman s request and continue indefinitely. A balanced perspective should be maintained. DES-exposed women who have a screen-detected abnormality should be managed in a specialist centre by an experienced colposcopist. Level IV (Hacker 2000, RCOG 2002) Level IV (Hacker 2000, RCOG 2002) Level IV (Hacker 2000, RCOG 2002 xviii Screening to prevent cervical cancer

22 1 Guidelines review process 1.1 Introduction A multidisciplinary guidelines review group was established to review and update the guidelines, including representatives of the Royal Australian and New Zealand College of Obstetricians and Gynaecologists, the Australian College of Rural and Remote Medicine, the Royal Australian College of General Practitioners, the Australian Society of Colposcopy and Cervical Pathology, the Royal College of Pathologists of Australasia, and the Australian Government Department of Health and Ageing. In addition, the review group included representatives with special expertise in cytology, virology, gynaecology, gynaecological oncology, epidemiology, health economics, health communication, Aboriginal and Torres Strait Islander women s health, and consumer concerns. A three-month public consultation process was undertaken, including an advertisement in The Australian newspaper inviting submissions. A website was developed as a vehicle for stakeholder information and consultation. Press releases informing stakeholders of the consultation process and inviting comment were forwarded to all appropriate professional bodies. Presentations were also given at professional meetings of these colleges, and two additional meetings were held for clinicians. Cervical screening programs in each state organised women s consultation sessions. Further consultations about the guidelines were also conducted in November and December 2003 (see Appendix 2). The recommendations of these guidelines are underpinned by an extensive review of the literature (see Appendixes 2 and 3). However, because of differences in terminology there have been some concerns about the applicability of many overseas studies to the Australian situation. Nevertheless, Australia is very fortunate because extensive case data on Pap tests is held in each state and territory Pap test register. Analysis of these data has been undertaken for the Guidelines Review Group by Dr Heather Mitchell with the assistance of each state and territory Pap test register. This has provided the Guidelines Review Group with a unique source of data upon which to base risk assessment and recommendations for follow-up. The Guidelines Review Group used the methods outlined in A Guide to the Development, Implementation and Evaluation of Clinical Practice Guidelines (NHMRC 1999) and the companying handbooks (NHMRC 2000abc, 2001, 2002) to review the literature and assess the evidence used to support these guidelines. 1.2 Literature review The Guidelines Review Group developed relevant clinical questions about the natural history of cervical cancer, the prognosis of women with different screen-detected abnormalities and the available options for management of cervical abnormalities and cervical cancer in Australia (see Appendix 3). Evidence from published studies was predominantly obtained from a thorough search of electronic databases. These included MEDLINE, EMBASE, the specialised trials register maintained by the Cochrane Gynaecological Cancer Group, the Cochrane Database of Systematic Reviews and the Cochrane Library, including the Database of Abstracts of Guidelines review process 1

23 Reviews of Effectiveness (DARE), Cancerlit, HealthSTAR and CINAHL. Additionally, potential relevant studies were identified from review articles and the references of identified studies, and unpublished studies were sought from the review group members. General search criteria included all studies published in the English language covering the period from 1966 onward. The exception to this search timeframe was for studies that related to Digene Hybrid Capture II. This test only became available in Different medical subject headings (MeSH) or search terms were used for different questions, but a general search strategy was used to identify systematic reviews and randomised controlled trials (see Appendix 3). Studies identified were checked against the clinical questions, and studies that answered the specific questions were included in the development of the guidelines. Articles that did not address the questions were excluded from review process. 1.3 Assessment and classification of evidence Studies identified from the literature for inclusion in the review and development of the guidelines were initially classified according to the NHMRC levels of evidence (Table 1.1). Where possible, the strength and quality of the evidence was further evaluated using the NHMRC-recommended strategy (NHMRC 2000b; see Table 1.2). Table 1.1 NHMRC levels of evidence I II III-1 III-2 III-3 IV Source: NHMRC 1999 Evidence obtained from a systematic review of all relevant randomised controlled trials. Evidence obtained from at least one properly designed randomised controlled trial. Evidence obtained from well-designed pseudo-randomised controlled trials (alternate allocation or some other method). Evidence obtained from comparative studies with concurrent controls and allocation not randomised (cohort studies), case control studies, or interrupted time series with a control group. Evidence obtained from comparative studies with historical control, two or more singlearm studies, or interrupted time series without a parallel control group. Evidence obtained from case series, either post-test or pretest and post-test. Table 1.2 Dimensions of evidence Dimension Strength of evidence: Level Quality Statistical precision Size of effect Relevance Source: NHMRC 2000b Definition Based on study design used, as an indicator of the degree to which bias has been eliminated by design. The methods used by investigators to minimise bias within a study design. The p-value or, alternatively, the precision of the estimate of the effect (as indicated by the confidence interval). It reflects the degree of certainty about the existence of true effect. The distance of the study estimate from the null value and the inclusion of only clinically important effects in the confidence interval. The usefulness of the evidence in clinical practice, particularly the appropriateness of the outcome measures used. 2 Screening to prevent cervical cancer

24 It is important to note that conducting randomised controlled trials in the area of cervical cancer management is not always practical and in many cases would be unethical. Consequently, there is very little evidence at Level I and II in this document. However, most of the management recommendations have been well tested through current clinical practice and many of the studies reviewed for this document were Level IV (case series). Experts in the field were also contacted to provide evidence based on their experience (whether published or not). Personal communication was also used as a proxy evidence measure. Although every effort was made to identify and obtain as many published articles as possible, for some questions, no articles or other clinical study information was found. In this situation, the broad agreement of peers on established best practice forms the best available basis for decision making, which can subsequently be evaluated. The Guidelines Review Group has therefore opted to include some recommendations based on a consensus of expert opinion within the review group about current best practice. Where this is the basis for a guideline, it is designated by consensus throughout this document. In addition, many of the clinical questions relevant for these guidelines are questions of prognosis. The National Cervical Screening Program has access to large population datasets through its state and territory Pap test registers. Using these registers, large-scale cohort studies of Australian women with Pap test-detected abnormalities have been conducted to compare outcomes over time from different test results and management options (see Appendixes 7 and 8). These large cohort studies, designated as Level III-2 under the NHMRC scheme (Table 1.1), represent high-quality evidence for answering prognosis questions (NHMRC 2000a) and are therefore a scientifically valid evidence base for many of the guidelines. Table 1.3 summarises the different categories of evidence used in these guidelines. Table 1.3 Designation of evidence used in these guidelines Evidence Clinical studies (randomised controlled trials, case control, cohort studies etc) Expert opinion of current best practice Data from Pap test registries Designation in guidelines Levels I, II etc (see Table 1.1) plus further details of strength and quality as appropriate (see Table 1.2) Consensus Australian registry data Guidelines review process 3

25

26 2 Cervical cancer in Australia Australia has the second lowest incidence of cervical cancer in the world among countries with comparable cancer registration systems (GLOBOCAN 2002). This is a remarkable achievement and is largely attributed to the successful implementation of the National Cervical Screening Program (NCSP), an organised approach to cervical screening in Australia implemented by a range of stakeholders, including general practitioners, gynaecologists, gynaecological oncologists, cytologists, women s health nurses, and national and state governments. Table 2.1 shows the Australian incidence and mortality rates for cervical cancer in comparison with some other countries for the period up to, and including, Table 2.1 Incidence of cervical cancer and mortality rate, selected countries, 2002 Country Incidence per 100,000 women (ASR) Mortality per 100,000 women (ASR) New Zealand United Kingdom Sweden United States Canada Australia Finland ASR = age standardised rate (World Standard Population) Source: GLOBOCAN Cervical screening became available for Australian women in the mid-1960s. The approach was opportunistic in that no formal policies were developed, there was no systematic recruitment of women, comprehensive quality control was difficult because of the fragmentation of women s tests across many laboratories, and there was no fail-safe system for treatment of women with abnormal cytology results. While some impact was made on the mortality from cervical cancer, the results were less than optimal. In 1991, an organised approach, the NCSP, was introduced as a national strategy to achieve further reductions in the incidence of and mortality from cervical cancer. The program has six major components: improving communication and education for women and health professionals establishing an infrastructure for a systematic approach to screening facilitating the regular participation of women in screening programs improving the quality control in smear taking and in laboratories processing and reporting cervical cytology instituting a fail-safe approach to the follow-up and management of screen-detected abnormalities monitoring and evaluating the organised approach. Cervical cancer in Australia 5

27 At around the same time, a two-yearly cervical screening interval was nationally recommended and introduced for the target population aged years. State and territory Pap test registries were developed as the infrastructure, and these have now matured into functional units that are integral to the success of the screening program. In 1987, there were 1092 new cases of cervical cancer in Australia. By 2000, this number had declined to 745 cases and, of these, 578 occurred in the target age group aged years (AIHW AACR 2003a). Overall, cervical cancer incidence and mortality have declined in the target age group by 56.9% and 57.7%, respectively, over the past decade (AIHW AACR 2003b). Figure 2.1 shows the incidence of cervical cancer in Australia by age in 1998, 1999 and New cases per 100,000 women Age group Note: Rates are expressed per 100,000 women; age standardised to the 2001 Australian Standard Population Source data: National Cancer Statistics Clearing House (AIHW-AACR 2003a, 2003b) Figure 2.1 Age-specific incidence rates of cervical cancer, Australia, There is some geographic variation within Australia in both incidence and mortality, but these differences are decreasing. During the period , the age-standardised incidence rate was 14.2 per 100,000 for metropolitan women, 13.6 per 100,000 for rural women and 15.9 per 100,000 for women in remote areas (AIHW AACR 2003a). During , the mortality rate was 2.1 per 100,000 for metropolitan women, 2.7 per 100,000 for rural women and 3.0 per 100,000 for remote-area women (AIHW AACR 2003a). Part of the excess incidence and mortality in remote Australia is attributed to the high disease rates among Indigenous women. Although these rates are decreasing, the risk of dying from cervical cancer for an Indigenous woman is still about six times that of a non-indigenous woman (AIHW AACR 2003a). In 1989, it was estimated that cervical cancer screening was preventing only 46% of squamous malignancies compared with a projected optimal prevention of 90%. By 1998, it was estimated that 70% of squamous cancers were being prevented again a remarkable testimony to the success of the NCSP (Mitchell 2003). 6 Screening to prevent cervical cancer

28 Figure 2.2 shows time trends between 1989 and 2000 in the incidence of the various histological types of cervical cancer among women aged years (AIHW 2004). 16 New cases per 100,000 women Squamous Adenocarcinoma Adeno-squamous Other Note: Age standardised to the 2001 Australian Standard Population Source: National Cancer Statistics Clearing House (AIHW 2004) Figure 2.2 Age-standardised incidence rates for cervical cancer, by histological type, for women aged years, Australia, Over the period shown in Figure 2.2, the incidence of squamous cervical cancer fell substantially, but the incidence of adenocarcinoma remained essentially unchanged. This is generally attributed to sampling difficulties in obtaining cells from the area where adenocarcinoma arises, problems in pathological interpretation and variations in clinical investigation and treatment. Cervical cancer in Australia 7

29

30 3 Current concepts of HPV infection and the natural history of cervical neoplasia 3.1 HPV infection causes cancer of the cervix There is overwhelming evidence that infection with human papillomavirus (HPV) is necessary, though not sufficient, for development of cancer of the cervix. Recent data, using sensitive methods to detect HPV, demonstrate that over 99.7% of cervical cancers test positive for HPV DNA (Bosch 2000). The attributable risk is greater than the risk of lung cancer conferred by smoking. The National Institutes of Health Consensus Conference on Cervical Cancer concluded cervical cancer is unique in that it is the first solid tumour to be shown to be virally induced in essentially every case (Braly 1996). 3.2 Cervical cancer is a rare outcome of HPV infection While persisting infection of the cervix with a high-risk HPV is necessary for the development of cervical cancer, it is certainly not sufficient (Walboomers et al 1999). Worldwide, the estimated prevalence of genital HPV infection is 326 million amongst adult women. This compares with an annual incidence of approximately 450,000 new cases of cervical cancer worldwide (Bosch 2000), indicating that cervical cancer is a rare outcome even of high-risk HPV infection. The modal age at first infection with high-risk genital HPV is between 15 and 25. The point prevalence among sexually active young women is high at around 20% to 25% (Ho et al 1998). Repeated testing of teenagers over a three-year period has documented a cumulative prevalence rate of 44% (95% CI, 40 to 48) (Woodman et al 2001). Infection may be with one or more HPV subtypes and these may change over time. In contrast, the modal age of diagnosis with cervical cancer in unscreened women varies between 35 and 50, depending on the country. Furthermore, less than 0.2% of cervical cancers occur in women under 25 (Burk et al 1996, Bosch and de Sanjose 2003, Sherman et al 2003a). These data suggest that progression of HPV infection to cancer is slow, and this is confirmed by longitudinal epidemiological studies (Burk et al 1996, Bosch and de Sanjose 2003, Sherman et al 2003a). The lifetime risk of cervical cancer, given infection with high-risk HPV, varies across geographical regions from 1 in 15 to 1 in 100 (Bosch and de Sanjose 2003, Schiffman and Kjaer 2003). Thus, screening for high-risk HPV infection would identify a very large number of women, only a few of whom are at risk of cervical cancer; the sensitivity of the test for risk is high, but the specificity is very low. Current concepts of HPV infection 9